Commit
•
60531cb
1
Parent(s):
1c96d88
Training in progress, step 2000
Browse files- .gitignore +1 -0
- config.json +83 -0
- pytorch_model.bin +3 -0
- training_args.bin +3 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
config.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"act_dim": 6,
|
| 3 |
+
"action_tanh": true,
|
| 4 |
+
"activation_function": "relu",
|
| 5 |
+
"adv_act_dim": 6,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"AdversarialDT"
|
| 8 |
+
],
|
| 9 |
+
"attn_pdrop": 0.1,
|
| 10 |
+
"bos_token_id": 50256,
|
| 11 |
+
"context_size": 20,
|
| 12 |
+
"embd_pdrop": 0.1,
|
| 13 |
+
"eos_token_id": 50256,
|
| 14 |
+
"flag": false,
|
| 15 |
+
"hidden_size": 128,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"lambda1": 1.0,
|
| 18 |
+
"lambda2": 1.0,
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"log_interval_steps": 100,
|
| 21 |
+
"max_ep_len": 1001,
|
| 22 |
+
"max_ep_return": 5389,
|
| 23 |
+
"max_obs_len": 1001,
|
| 24 |
+
"max_obs_return": 5389,
|
| 25 |
+
"min_ep_return": -36.38545789499999,
|
| 26 |
+
"min_obs_return": -36.38545789499999,
|
| 27 |
+
"model_type": "decision_transformer",
|
| 28 |
+
"n_head": 1,
|
| 29 |
+
"n_inner": null,
|
| 30 |
+
"n_layer": 3,
|
| 31 |
+
"n_positions": 1024,
|
| 32 |
+
"pr_act_dim": 6,
|
| 33 |
+
"reorder_and_upcast_attn": false,
|
| 34 |
+
"resid_pdrop": 0.1,
|
| 35 |
+
"returns_scale": 1000,
|
| 36 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 37 |
+
"scale_attn_weights": true,
|
| 38 |
+
"state_dim": 17,
|
| 39 |
+
"state_mean": [
|
| 40 |
+
-0.12463613730660446,
|
| 41 |
+
0.0380851573126452,
|
| 42 |
+
0.15752106892811954,
|
| 43 |
+
0.09558710425767158,
|
| 44 |
+
0.07873622143234135,
|
| 45 |
+
-0.024131525166433174,
|
| 46 |
+
-0.12270041477603931,
|
| 47 |
+
-0.2951165516992248,
|
| 48 |
+
3.0499440021372095,
|
| 49 |
+
-0.025617095212702615,
|
| 50 |
+
-0.03129052119123046,
|
| 51 |
+
-0.008744465966597231,
|
| 52 |
+
-0.08879571539228269,
|
| 53 |
+
0.126699288294097,
|
| 54 |
+
-0.08706499813665808,
|
| 55 |
+
-0.2625399151472837,
|
| 56 |
+
-0.20975868968367148
|
| 57 |
+
],
|
| 58 |
+
"state_std": [
|
| 59 |
+
0.08693300335893665,
|
| 60 |
+
0.25115436008843744,
|
| 61 |
+
0.3412246248710908,
|
| 62 |
+
0.37199314748513224,
|
| 63 |
+
0.4248907200422569,
|
| 64 |
+
0.4394592313301521,
|
| 65 |
+
0.30854268429829035,
|
| 66 |
+
0.2503472104454028,
|
| 67 |
+
1.9948001599895642,
|
| 68 |
+
0.8866457906518544,
|
| 69 |
+
1.4982192253144506,
|
| 70 |
+
7.182257934162716,
|
| 71 |
+
7.8703290304262445,
|
| 72 |
+
9.443924738229082,
|
| 73 |
+
9.063740430833478,
|
| 74 |
+
6.143162458510403,
|
| 75 |
+
5.662356308267215
|
| 76 |
+
],
|
| 77 |
+
"torch_dtype": "float32",
|
| 78 |
+
"total_train_steps": 10000,
|
| 79 |
+
"transformers_version": "4.29.2",
|
| 80 |
+
"use_cache": true,
|
| 81 |
+
"vocab_size": 1,
|
| 82 |
+
"warmup_steps": 1000
|
| 83 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5a5b772bc01beb5c47c09ad1e2c5e60f2e157c6a6eab73283e02736f9f9d7b6
|
| 3 |
+
size 6625853
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6583c7b9b926da33c750c69a636dd592efd018652e8997adc9c53ebf15c28ecd
|
| 3 |
+
size 4664
|